What Technology Changes Will Affect Your Practice Soon?

James M. Lebret, MD


February 01, 2017

In This Article

Technologies That Are at Least 2 Years Off

Researchers and companies are working hard on a number of new technologies, but frontline doctors are unlikely to see them because they will work behind the scenes of digital systems. Expect more uses in medicine over the next 2 years.

Artificial intelligence (AI). AI includes natural language processing, machine vision, and machine learning. This type of software is increasingly able to perform actions formerly ascribed only to humans, such as understanding voice and text and independently learning new tasks and solving problems.

For example, a patient taking anticoagulation medication who is monitored by a wireless ECG monitor for atrial fibrillation may develop cerebral T waves in response to an intracranial hemorrhage. It would be a liability to have the ECG data without the manpower to identify the new change. But with wireless ECGs expected to become ubiquitous, not even an army of clinicians could comb through the deluge of data that will result.

This is where AI tools such as machine vision and machine learning can help. These tools learn to act in concert to detect the T waves, check them against a compendium of pathophysiologic waveforms, and alert the clinician to the possible bleed.

Such a system is a quantum leap beyond basic threshold alerts of telemetry systems. These tools, which mimic human cognition and have the potential to open the data bottleneck, are necessary to offload some of the manual chart review of exponentially increasing patient data and can literally help save lives.

Predictive analytics. Overlapping AI, predictive analytics refers to the software businesses use to analyze trends and to make predictions about the future. Think of them as clinical prediction rules, such as the thrombolysis in myocardial infarction (TIMI) score for non-ST-segment myocardial infarction (NSTEMI)—but much more powerful.

As software dashboards become more commonplace, expect to see them show predictions about the probability that a patient with chronic obstructive pulmonary disease will be readmitted or how a patient's blood pressure may respond to an angiotensin-converting enzyme inhibitor. Similarly, you might receive an alert about a patient who is anticipated to require intensive care unit admission in the next 36 hours.

While great clinicians often have uncanny insight into which patient is sickest or what the best treatment is, not every patient is lucky enough to have an experienced doctor at the bedside at all times. Predictive analytics provide physician-level insights during nights and weekends.

As these technologies mature, they will help guide the clinician's medical decision-making, similar to how more recently adopted technologies, such as CT scans and EHRs, do today. While these technologies will never replace clinicians, they will help deliver a high level of patient care.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.